liveBook · Manning

Parameter

As explained in AI Governance

A model parameter is a setting that controls a model's behavior during generation or operation.

A model parameter is a setting that affects generation. Lowering temperature can make answers more consistent, while reducing creative range; these controls do not guarantee full determinism. section 6.1.3

For more stable answers:

  • Set temperature to zero to favor the single most likely next word. section 6.1.3
  • Lower temperature and fix the random seed to increase consistency, while recognizing that outputs may still not be fully deterministic. section 6.1.3

How AI Governance covers it

How deep it goes
Specifies temperature and random-seed adjustments, their creative-range trade-off, and their limit short of full determinism.
Examples
Uses a zero-temperature setting and a fixed random seed as consistency examples.
What you can do after
Configure model parameters to improve consistency while accounting for reduced creative range and remaining nondeterminism.
How it connects
Links temperature and random seed settings with prompt structure and model-version choices.

Also explained in

Compare how 41 books explain parameter →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI Governance. The section links point to the parts of the book it draws on.